Program introduces students to machine learning and security research

Jessica Hallman
August 10, 2020

UNIVERSITY PARK, Pa. — This summer, three undergraduate students from three higher education institutions got an exclusive, in-depth introduction to research topics focused on machine learning in cybersecurity through the Research Experiences for Undergraduates (REU) site program sponsored by National Science of Foundation (NSF) and hosted by Penn State’s College of Information Sciences and Technology.

Through the intensive 10-week program, the undergraduates were matched 1:1 with faculty experts at the College of IST to focus on their selected research activities. The program was designed to encourage research-oriented undergraduate students to explore continued opportunities through graduate programs.

“A lot of undergraduate students don’t know that research is an option they can [pursue] in the future,” said Aiping Xiong, co-principal investigator of the REU site program. “That’s why we want to make it accessible.”

Sophia Hager, a junior studying computer science at Smith College in Northampton, Massachusetts, was one of the students selected for the program. Along with her Penn State faculty mentor Dongwon Lee, associate professor of information sciences and technology, Hager spent the summer developing a demo website that uses a neural net — a set of algorithms developed to identify patterns — for users to check online articles for healthcare misinformation.

“What I’ve found most interesting is seeing how neural nets are used in projects like this, since a lot of my experience has come from very theoretical exercises,” said Hager. “One of the things I’ve noticed [during the] pandemic is how rapidly misinformation can spread, and working on this project has given me much more of an understanding of how using machine learning in fake news detection is relevant and useful in a non-academic setting.”

Hager said that her experience with the REU site program has inspired her to pursue graduate school in the future.

Liam Boston, a junior studying information sciences and technology at Penn State, had the opportunity to strengthen his IST background through the program. He spent the summer working with Benjamin Hanrahan to develop a plugin for Microsoft Word. The app takes a rudimentary approach to train users in basic Word skills, such as bolding certain words or correcting spelling errors.

For Boston, the REU site program served as an extension of what he has learned through the curriculum at Penn State.

“The 1:1 nature of the REU is a much more hands-on learning experience; it’s less ‘reading a book’ and more ‘learning on your own,’” said Boston. “In class it’s hard to visualize how different topics and skills can actually be used, and this program was a great opportunity for me to apply what I know while also picking up new skills along the way.”

Rachael He, a junior studying computer science at the University of Rochester in Rochester, New York, also participated. Together with her Penn State faculty mentor Aiping Xiong, assistant professor of information sciences and technology, He spent the summer developing a human subject experiment about explainable artificial intelligence in the misinformation domain. Specifically, the experiment examined the impacts of counterfactually feature-based explanations on people’s beliefs in news headlines on social media.

“I learned a lot about AI and particularly about XAI [explainable AI], along with fake news and psychology,” said He. “XAI has always been a topic I've been interested in, even if I didn't know there was a name for the area at the time. Thus, I definitely want to pursue the topic in the future.”

Funded by the National Science Foundation, the Machine Learning in Cybersecurity program is the first REU site program hosted by the College of IST. Lee, who is the principal investigator for the program, hopes to host additional opportunities in the future.

“We are trying to attract [the student participants] to academia, which will not only graduate programs but is beneficial to industry too,” said Lee. “What students learn is not just to write a research article, but the exploratory mindset. They learn how to identify a problem and come up with a plan for how to solve it.”

Originally, nine students from institutions across the country were recruited to participate in a physical summer program. Due to the novel coronavirus pandemic and restrictions on physical summer activities at Penn State, a pilot virtual program involving three students was held instead at the recommendation of the NSF.

Last Updated June 28, 2021